1 Mon Jan 15 11:03:20 2007 Ben Pfaff <blp@gnu.org>
3 Fix bugs found by valgrind when --enable-debug is used with the
4 new case code. These bugs are hidden when the data set is small
5 enough to find in memory; if a bigger data set that would overflow
6 to disk were used, then data corruption would occur.
8 * chisquare.c (create_freq_hash): Pass free_freq_mutable_hash to
9 hsh_create as free function. Make copy of data put into hash.
11 * oneway.q (free_value): New function.
12 (run_oneway): Use free_value as arg to hsh_create. Make copy of
15 * rank.q (rank_cases): Don't access data in case after we've given
18 Tue Jan 9 19:16:11 2007 Ben Pfaff <blp@gnu.org>
22 * aggregate.c (parse_aggregate_functions) Initialize
25 Fri Dec 22 14:04:09 2006 Ben Pfaff <blp@gnu.org>
27 Simplify missing value handling.
29 * aggregate.c (struct agr_var): Remove `bool include_missing', add
30 `enum mv_class exclude'. Remove `int missing', add `bool
31 saw_missing'. Update users.
33 * descriptives.c (struct dsc_trns): Removed `int
34 include_user_missing', add `enum mv_class exclude'. Update users.
35 (struct dsc_proc): Ditto.
37 * examine.q: (static var value_is_missing): Rename
38 `exclude_values', change type to `enum mv_class'. Update users.
42 Fri Dec 22 19:22:18 WST 2006 John Darrington <john@darrington.wattle.id.au>
44 * frequencies.q : Fixed bug #17420, where the table bounds were overun
45 when /FORMAT=nolabels was given.
47 Wed Dec 20 18:45:31 WST 2006 John Darrington <john@darrington.wattle.id.au>
49 * binomial.c binomial.h : New files. Thanks to Jason Stover
50 for assistance with these.
52 * chisquare.c chisquare.h freq.c freq.h npar-summary.c npar-summary.h
53 npar.h npar.q: New files. Implementing NPAR TESTS.
55 * frequencies.q : Moved structure definitions into freq.[ch]
57 Sat Dec 16 22:26:44 2006 Ben Pfaff <blp@gnu.org>
59 Make it possible to pull cases from the active file with a
60 function call, instead of requiring indirection through a callback
63 * aggregate.c (cmd_aggregate): Take advantage of new procedure
65 (agr_to_active_file): Removed.
66 (presorted_agr_to_sysfile): Removed.
68 * autorecode.c (cmd_autorecode): Take advantage of new procedure
70 (autorecode_proc_func): Removed.
72 * flip.c (struct flip_pgm): New members to allow conformance with
73 new case_source_class interface.
74 (cmd_flip): Adapt to new case_source_class interface.
75 (flip_source_read): Ditto.
76 (flip_source_destroy): Ditto.
78 Sat Dec 16 12:54:27 2006 Ben Pfaff <blp@gnu.org>
80 * rank.q (rank_custom_variables): Allow grouping variables to be
81 strings. Fixes bug #18533. Thanks to John Darrington for review.
83 Sat Dec 9 18:47:51 2006 Ben Pfaff <blp@gnu.org>
85 * regression.q (is_depvar): Compare variable pointers instead of
88 Thu Dec 7 15:26:25 WST 2006 John Darrington <john@darrington.wattle.id.au>
90 * examine.q: Allocated the categorical values for the dependent and
91 independent variables, on the heap. Hence they can be of any width.
93 Wed Dec 6 21:14:26 2006 Ben Pfaff <blp@gnu.org>
95 * regression.q (reg_inserted): Compare variable pointers instead
98 Mon Dec 4 22:33:46 2006 Ben Pfaff <blp@gnu.org>
100 * crosstabs.q (insert_summary): Use var_to_string for labeling.
101 (output_pivot_table) Ditto.
104 * frequencies.q (setup_z_trns): Ditto.
106 (dump_condensed) Ditto.
107 (dump_statistics) Ditto.
110 Sun Nov 5 08:31:42 WST 2006 John Darrington <john@darrington.wattle.id.au>
112 * t-test.q, oneway.q: Changed to use the new casefilter structure.
114 Sat Oct 14 16:52:28 2006 Ben Pfaff <blp@gnu.org>
116 * rank.q: (rank_sorted_casefile) Add some missing case_destroy()
117 calls to fix a memory leak.
119 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
121 * rank.q: Plugged a small memory leak which occurred under error
124 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
126 * rank.q: Implemented most of the RANK command.
128 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
130 * regression.q (run_regression): New function to move knowledge of
131 pspp_linreg_cache out of math/coefficient.[ch].
133 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
135 Fix bug #11612, "q2c documentation does not agree with code".
137 * examine.q: Audit use of q2c "+" prefixes that indicate that a
138 command may appear multiple times.
140 * frequencies.q: Ditto.
144 * regression.q: Ditto.
148 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
150 Support long string variables on FREQUENCIES, as
151 an extension when in enhanced algorithms mode. For Greg Hunt
152 <greg@firmansyah.com>.
154 * frequencies.q: (struct freq) Change `v' member from union value
155 to union value *. Update all references.
156 (struct var_freqs) Add width, print members to represent effective
157 variable width and display format.
158 (calc) Copy entire long string value into the hash table.
159 (frq_custom_variables) Set new width, print members.
160 (hash_value_alpha) Get width from var_freqs.
161 (compare_value_alpha_a) Ditto.
162 (compare_freq_alpha_a) Ditto.
163 (compare_freq_alpha_d) Ditto.
164 (dump_full) Get display format from var_freqs.
165 (dump_condensed) Ditto.
167 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
169 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
170 bytes of string variables into the output cells, seeing as we only
172 (dump_condensed) Ditto.
174 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
176 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
178 * frequencies.q: (hsh_hash_bytes) We only copy the first
179 MAX_SHORT_STRING bytes of string variables, so we must only
180 compare that many bytes, even if the string variable is longer.
181 (compare_value_alpha_a) Ditto.
182 (compare_freq_alpha_a) Ditto.
183 (compare_freq_alpha_d) Ditto.
185 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
187 * regression.q: Adjusted code to account for cache->coeff being a
188 pspp_linreg_coeff **.
190 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
194 * aggregate.c (cmd_aggregate): Free default_dict before replacing
197 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
199 * flip.c (flip_file): Check return value of pool_fclose().
201 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
203 Get rid of `char *c' member in union value, for cleanliness.
205 * aggregate.c: (union agr_argument) New union.
206 (struct agr_var) Change element type of arg[] from union value to
208 (parse_aggregate_functions) Change local variable types likewise.
210 * autorecode.c: (union arc_value) New union.
211 (struct arc_item) Change "from" from union value to union
213 (recode) Change local variable from union value to union
215 (autorecode_trns_proc) Ditto.
216 (compare_alpha_value) Ditto.
217 (hash_alpha_value) Ditto.
218 (compare_numeric_value) Ditto.
219 (hash_numeric_value) Ditto.
220 (autorecode_proc_func) Ditto.
222 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
224 Continue reforming procedure execution. In this phase, get rid of
225 the output code for SPLIT FILE groups in procedure.c, which really
226 shouldn't be doing any output. Move it into the individual
227 procedures instead. This also adds some flexibility.
229 * crosstabs.q (precalc): Call output_split_file_values().
231 * descriptives.c (calc_descriptives): Ditto.
233 * examine.q (run_examine): Ditto.
235 * frequencies.q (precalc): Ditto.
237 * oneway.q (run_oneway): Ditto.
239 * regression.q (run_regression): Ditto.
241 * t-test.q (calculate): Ditto.
243 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
245 Continue reforming procedure execution. In this phase, get rid of
246 many global variables, consolidating procedure execution in
247 procedure.c. Encapsulate transformations in new "struct
248 trns_chain". Also, change implementation of N OF CASES, FILTER,
249 and PROCESS IF from special cases to transformations.
251 * aggregate.c (cmd_aggregate) Use discard_variables().
253 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
255 * regression.q (regression_trns_resid_proc): Pass only the
256 variables used in the model to (*model->residual)().
258 * regression.q (regression_trns_pred_proc): Pass only the
259 variables used in the model to (*model->pred)().
261 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
263 * regression.q: Added support for multiple transformations.
265 * regression.q (regression_trns_resid_proc): New function.
267 * regression.q (regression_trns_pred_proc): New function.
269 * regression.q (subcommand_save): Added support for saving
272 * regression.q (regression_trns_free): New function.
274 * regression.q (reg_get_name): New function.
276 * regression.q (reg_save_var): New function.
278 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
280 * rank.q (parse_rank_function): Use SE instead of ME for parse
283 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
285 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
288 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
290 * regression.q (try_name): New function. (Partly copied from
291 try_name in descriptives.c.)
293 * regression.q (subcommand_save): Choose residual variable names
296 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
298 * regression.q (cmd_regression): Moved call to subcommand_save()
299 outside multipass_procedure_with_splits().
301 * regression.q (regression_trns_proc): Fixed value counter n_vals
302 before calling *model->residual().
304 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
306 * regression.q (regression_trns_proc): Fixed the look-up of the
309 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
311 * regression.q (regression_trns_proc): Look up the residual
312 variable in the linear regression cache.
314 * regression.q (subcommand_save): Set the residual variable in the
315 linear regression cache.
317 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
319 * regression.q (regression_trns_proc): Accept case_idx as an int
320 to match the definition of trns_proc_func.
322 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
324 * regression.q (regression_trns_proc): New function.
326 * regression.q (subcommand_save): Create variable residuals and
327 add a transformation to assign values to them. Also free the
328 linreg_cache if the SAVE command was not called. Removed the
329 casereading loop. Placed actual computation of residuals in
330 regression_trns_proc.
332 * regression.q (run_regression): Moved call to free
333 pspp_linreg_cache to subcommand_save.
335 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
337 * examine.q (output_examine): Add casts to fix warnings.
339 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
341 * regression.q (subcommand_save): New function.
343 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
345 * regression.q: New function reg_has_categorical () to tell
346 whether a pspp_linreg_struct was made with any variables of type
349 * regression.q: (subcommand_export): Call
350 reg_print_categorical_encoding() only if the model uses any
351 categorical variables.
353 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
355 * crosstabs.q: (output_pivot_table) Drop spurious space from
358 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
360 * regression.q: Added custom syntax parser for VARIABLES subcommand
362 * regression.q: Moved most instructions for run_regression ()
363 inside the loop over dependent variables.
365 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
367 * Moved files from src directory